Pith. sign in

REVIEW 6 cited by

Estimate-Then-Optimize versus Integrated-Estimation-Optimization versus Sample Average Approximation: A Stochastic Dominance Perspective

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2304.06833 v4 pith:T3L5SCNW submitted 2023-04-13 stat.ML cs.LGstat.ME

Estimate-Then-Optimize versus Integrated-Estimation-Optimization versus Sample Average Approximation: A Stochastic Dominance Perspective

classification stat.ML cs.LGstat.ME
keywords modelwhenoptimizationclassstochasticregretapproximationaverage
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

In data-driven stochastic optimization, model parameters of the underlying distribution need to be estimated from data in addition to the optimization task. Recent literature considers integrating the estimation and optimization processes by selecting model parameters that lead to the best empirical objective performance. This integrated approach, which we call integrated-estimation-optimization (IEO), can be readily shown to outperform simple estimate-then-optimize (ETO) when the model is misspecified. In this paper, we show that a reverse behavior appears when the model class is well-specified and there is sufficient data. Specifically, for a general class of nonlinear stochastic optimization problems, we show that simple ETO outperforms IEO asymptotically when the model class covers the ground truth, in the strong sense of stochastic dominance of the regret. Namely, the entire distribution of the regret, not only its mean or other moments, is always better for ETO compared to IEO. Our results also apply to constrained, contextual optimization problems where the decision depends on observed features. Whenever applicable, we also demonstrate how standard sample average approximation (SAA) performs the worst when the model class is well-specified in terms of regret, and best when it is misspecified. Finally, we provide experimental results to support our theoretical comparisons and illustrate when our insights hold in finite-sample regimes and under various degrees of misspecification.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Solver-Free Training Method for Predict-then-Optimize

    stat.ML 2026-06 unverdicted novelty 7.0

    Introduces a measure-transformation-based surrogate loss for solver-free training in predict-then-optimize problems, with Fisher consistency and excess risk bounds.

  2. Conformal Risk-Averse Decision Making with Action Conditional Guarantee

    stat.ML 2026-06 unverdicted novelty 7.0

    Action-conditional conformal prediction sets provide per-action safety guarantees for risk-averse policies that optimize conditional value-at-risk through pinball-loss minimization.

  3. Risk-Controlled Post-Processing of Decision Policies

    stat.ML 2026-05 unverdicted novelty 7.0

    Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i...

  4. Scaling Decision-Focused Learning to Large Problems with Lagrangian Decomposition

    cs.LG 2026-06 unverdicted novelty 6.0

    Lagrangian decomposition yields a scalable surrogate objective and losses for decision-focused learning that outperforms prior DFL methods on large multi-dimensional knapsack and quadratic portfolio instances.

  5. Risk-averse Decision Making with Contextual Information: Model, Sample Average Approximation, and Kernelization

    math.OC 2025-02 unverdicted novelty 4.0

    Establishes equivalence conditions between nested and joint risk assessments in contextual optimization, shows policy independence from contextual risk measure under conditions, and proves SAA consistency in RKHS.

  6. Decision-Focused Learning: When and Why Traditional Prediction Models Fail

    cs.LG 2026-06 unverdicted novelty 2.0

    A tutorial reviewing why traditional prediction models often fail to improve decision quality in stochastic optimization and summarizing key properties and tools of decision-focused learning.